A Channel-Dimensional Feature-Reconstructed Deep Learning Model for Predicting Breast Cancer Molecular Subtypes on Overall b-Value Diffusion-Weighted MRI.

BACKGROUND: Dynamic contrast-enhanced (DCE) MRI commonly outperforms diffusion-weighted (DW) MRI in breast cancer discrimination. However, the side effects of contrast agents limit the use of DCE-MRI, particularly in patients with chronic kidney disease. PURPOSE: To develop a novel deep learning model to fully exploit the potential of overall b-value DW-MRI without the need for a contrast agent in predicting breast cancer molecular subtypes and to evaluate its performance in comparison with DCE-MRI. STUDY TYPE: Prospective. SUBJECTS: 486 female breast cancer patients (training/validation/test: 64%/16%/20%). FIELD STRENGTH/SEQUENCE: 3.0 T/DW-MRI (13 b-values) and DCE-MRI (one precontrast and five postcontrast phases). ASSESSMENT: The breast cancers were divided into four subtypes: luminal A, luminal B, HER2+, and triple negative. A channel-dimensional feature-reconstructed (CDFR) deep neural network (DNN) was proposed to predict these subtypes using pathological diagnosis as the reference standard. Additionally, a non-CDFR DNN (NCDFR-DNN) was built for comparative purposes. A mixture ensemble DNN (ME-DNN) integrating two CDFR-DNNs was constructed to identify subtypes on multiparametric MRI (MP-MRI) combing DW-MRI and DCE-MRI. STATISTICAL TESTS: Model performance was evaluated using accuracy, sensitivity, specificity, and area under the receiver operating characteristic curve (AUC). Model comparisons were performed using the one-way analysis of variance with least significant difference post hoc test and the DeLong test. P < 0.05 was considered significant. RESULTS: The CDFR-DNN (accuracies, 0.79 ~ 0.80; AUCs, 0.93 ~ 0.94) demonstrated significantly improved predictive performance than the NCDFR-DNN (accuracies, 0.76 ~ 0.78; AUCs, 0.92 ~ 0.93) on DW-MRI. Utilizing the CDFR-DNN, DW-MRI attained the predictive performance equal (P = 0.065 ~ 1.000) to DCE-MRI (accuracies, 0.79 ~ 0.80; AUCs, 0.93 ~ 0.95). The predictive performance of the ME-DNN on MP-MRI (accuracies, 0.85 ~ 0.87; AUCs, 0.96 ~ 0.97) was superior to those of both the CDFR-DNN and NCDFR-DNN on either DW-MRI or DCE-MRI. DATA CONCLUSION: The CDFR-DNN enabled overall b-value DW-MRI to achieve the predictive performance comparable to DCE-MRI. MP-MRI outperformed DW-MRI and DCE-MRI in subtype prediction. LEVEL OF EVIDENCE: 2 TECHNICAL EFFICACY STAGE: 1.

Deep wavelet scattering orthogonal fusion network for glioma IDH mutation status prediction.

Accurately predicting the isocitrate dehydrogenase (IDH) mutation status of gliomas is greatly significant for formulating appropriate treatment plans and evaluating the prognoses of gliomas. Although existing studies can accurately predict the IDH mutation status of gliomas based on multimodal magnetic resonance (MR) images and machine learning methods, most of these methods cannot fully explore multimodal information and effectively predict IDH status for datasets acquired from multiple centers. To address this issue, a novel wavelet scattering (WS)-based orthogonal fusion network (WSOFNet) was proposed in this work to predict the IDH mutation status of gliomas from multiple centers. First, transformation-invariant features were extracted from multimodal MR images with a WS network, and then the multimodal WS features were used instead of the original images as the inputs of WSOFNet and were fully fused through an adaptive multimodal feature fusion module (AMF2M) and an orthogonal projection module (OPM). Finally, the fused features were input into a fully connected classifier to predict IDH mutation status. In addition, to achieve improved prediction accuracy, four auxiliary losses were also used in the feature extraction modules. The comparison results showed that the prediction area under the curve (AUC) of WSOFNet on a single-center dataset was 0.9966 and that on a multicenter dataset was approximately 0.9655, which was at least 3.9% higher than that of state-of-the-art methods. Moreover, the ablation experimental results also proved that the adaptive multimodal feature fusion strategy based on orthogonal projection could effectively improve the prediction performance of the model, especially for an external validation dataset.

Comparison of Dynamic Contrast-Enhanced MRI and Non-Mono-Exponential Model-Based Diffusion-Weighted Imaging for the Prediction of Prognostic Biomarkers and Molecular Subtypes of Breast Cancer Based on Radiomics.

BACKGROUND: Dynamic contrast-enhanced (DCE) MRI and non-mono-exponential model-based diffusion-weighted imaging (NME-DWI) that does not require contrast agent can both characterize breast cancer. However, which technique is superior remains unclear. PURPOSE: To compare the performances of DCE-MRI, NME-DWI and their combination as multiparametric MRI (MP-MRI) in the prediction of breast cancer prognostic biomarkers and molecular subtypes based on radiomics. STUDY TYPE: Prospective. POPULATION: A total of 477 female patients with 483 breast cancers (5-fold cross-validation: training/validation, 80%/20%). FIELD STRENGTH/SEQUENCE: A 3.0 T/DCE-MRI (6 dynamic frames) and NME-DWI (13 b values). ASSESSMENT: After data preprocessing, high-throughput features were extracted from each tumor volume of interest, and optimal features were selected using recursive feature elimination method. To identify ER+ vs. ER-, PR+ vs. PR-, HER2+ vs. HER2-, Ki-67+ vs. Ki-67-, luminal A/B vs. nonluminal A/B, and triple negative (TN) vs. non-TN, the following models were implemented: random forest, adaptive boosting, support vector machine, linear discriminant analysis, and logistic regression. STATISTICAL TESTS: Student's t, chi-square, and Fisher's exact tests were applied on clinical characteristics to confirm whether significant differences exist between different statuses (±) of prognostic biomarkers or molecular subtypes. The model performances were compared between the DCE-MRI, NME-DWI, and MP-MRI datasets using the area under the receiver-operating characteristic curve (AUC) and the DeLong test. P < 0.05 was considered significant. RESULTS: With few exceptions, no significant differences (P = 0.062-0.984) were observed in the AUCs of models for six classification tasks between the DCE-MRI (AUC = 0.62-0.87) and NME-DWI (AUC = 0.62-0.91) datasets, while the model performances on the two imaging datasets were significantly poorer than on the MP-MRI dataset (AUC = 0.68-0.93). Additionally, the random forest and adaptive boosting models (AUC = 0.62-0.93) outperformed other three models (AUC = 0.62-0.90). DATA CONCLUSION: NME-DWI was comparable with DCE-MRI in predictive performance and could be used as an alternative technique. Besides, MP-MRI demonstrated significantly higher AUCs than either DCE-MRI or NME-DWI. EVIDENCE LEVEL: 2. TECHNICAL EFFICACY: Stage 2.

Radiomic analysis based on multi-phase magnetic resonance imaging to predict preoperatively microvascular invasion in hepatocellular carcinoma.

BACKGROUND: The prognosis of hepatocellular carcinoma (HCC) remains poor and relapse occurs in more than half of patients within 2 years after hepatectomy. In terms of recent studies, microvascular invasion (MVI) is one of the potential predictors of recurrence. Accurate preoperative prediction of MVI is potentially beneficial to the optimization of treatment planning. AIM: To develop a radiomic analysis model based on pre-operative magnetic resonance imaging (MRI) data to predict MVI in HCC. METHODS: A total of 113 patients recruited to this study have been diagnosed as having HCC with histological confirmation, among whom 73 were found to have MVI and 40 were not. All the patients received preoperative examination by Gd-enhanced MRI and then curative hepatectomy. We manually delineated the tumor lesion on the largest cross-sectional area of the tumor and the adjacent two images on MRI, namely, the regions of interest. Quantitative analyses included most discriminant factors (MDFs) developed using linear discriminant analysis algorithm and histogram analysis with MaZda software. Independent significant variables of clinical and radiological features and MDFs for the prediction of MVI were estimated and a discriminant model was established by univariate and multivariate logistic regression analysis. Prediction ability of the above-mentioned parameters or model was then evaluated by receiver operating characteristic (ROC) curve analysis. Five-fold cross-validation was also applied via R software. RESULTS: The area under the ROC curve (AUC) of the MDF (0.77-0.85) outperformed that of histogram parameters (0.51-0.74). After multivariate analysis, MDF values of the arterial and portal venous phase, and peritumoral hypointensity in the hepatobiliary phase were identified to be independent predictors of MVI (P < 0.05). The AUC value of the model was 0.939 [95% confidence interval (CI): 0.893-0.984, standard error: 0.023]. The result of internal five-fold cross-validation (AUC: 0.912, 95%CI: 0.841-0.959, standard error: 0.0298) also showed favorable predictive efficacy. CONCLUSION: Noninvasive MRI radiomic model based on MDF values and imaging biomarkers may be useful to make preoperative prediction of MVI in patients with primary HCC.

An Unsupervised Deep Learning Approach for Dynamic-Exponential Intravoxel Incoherent Motion MRI Modeling and Parameter Estimation in the Liver.

BACKGROUND: Dynamic-exponential intravoxel incoherent motion (IVIM) imaging is a potential technique for prediction, monitoring, and differential diagnosis of hepatic diseases, especially liver tumors. However, the use of such technique at voxel level is still limited. PURPOSE: To develop an unsupervised deep learning approach for voxel-wise dynamic-exponential IVIM modeling and parameter estimation in the liver. STUDY TYPE: Prospective. POPULATION: Ten healthy subjects (4 males; age 28 ± 6 years). FIELD STRENGTH/SEQUENCE: Single-shot spin-echo echo planar imaging (SE-EPI) sequence with monopolar diffusion-encoding gradients (12 b-values, 0-800 seconds/mm2 ) at 3.0 T. ASSESSMENT: The proposed deep neural network (DNN) was separately trained on simulated and in vivo hepatic IVIM datasets. The trained networks were compared to the approach combining least squares with Akaike information criterion (LSQ-AIC) in terms of dynamic-exponential modeling accuracy, inter-subject coefficients of variation (CVs), and fitting residuals on the simulated subsets and regions of interest (ROIs) in the left and right liver lobes. The ROIs were delineated by a radiologist (H.-X.Z.) with 7 years of experience in MRI reading. STATISTICAL TESTS: Comparisons between approaches were performed with a paired t-test (normality) or a Wilcoxon rank-sum test (nonnormality). P < 0.05 was considered statistically significant. RESULTS: In simulations, DNN gave significantly higher accuracy (91.6%-95.5%) for identification of bi-exponential decays with respect to LSQ-AIC (79.7%-86.8%). For tri-exponential identification, DNN was also superior to LSQ-AIC despite not reaching a significant level (P = 0.08). Additionally, DNN always yielded comparatively low root-mean-square error for estimated parameters. For the in vivo IVIM measurements, inter-subject CVs (0.011-0.150) of DNN were significantly smaller than those (0.049-0.573) of LSQ-AIC. Concerning fitting residuals, there was no significant difference between the two approaches (P = 0.56 and 0.76) in both the simulated and in vivo studies. DATA CONCLUSION: The proposed DNN is recommended for accurate and robust dynamic-exponential modeling and parameter estimation in hepatic IVIM imaging. LEVEL OF EVIDENCE: 2 TECHNICAL EFFICACY: Stage 1.

Non-mono-exponential diffusion models for assessing early response of liver metastases to chemotherapy in colorectal Cancer.

BACKGROUND: Preoperative chemotherapy is becoming standard therapy for liver metastasis from colorectal cancer, so early assessment of treatment response is crucial to make a reasonable therapeutic regimen and avoid overtreatment, especially for patients with severe side effects. The role of three non-mono-exponential diffusion models, such as the kurtosis model, the stretched exponential model and the statistical model, were explored in this study to early assess the response to chemotherapy in patients with liver metastasis from colorectal cancer. METHODS: Thirty-three patients diagnosed as colorectal liver metastasis were evaluated in this study. Diffusion-weighted images with b values (0, 200, 500, 1000, 1500, 2000 s/mm2) were acquired at 3.0 T. The parameters (ADCk, K, DDC,α, Ds and σ) were derived from three non-mono-exponential models (the kurtosis, stretched exponential and statistical models) as well as their corresponding percentage changes before and after chemotherapy. The difference in above parameters between the response and non-response groups were analyzed with independent-samples T-test (normality) and Mann-Whitney U-test (non-normality). Meanwhile, receiver operating characteristic curve (ROC) analyses were performed to assess the response to chemotherapy. RESULTS: Significantly lower values of K (the kurtosis coefficient derived from the kurtosis model) and σ (the width of diffusion coefficient distribution in the statistical model) (P < 0.05) were observed in the respond group before treatment, as well as higher ΔK and Δσ values (P < 0.05) after the first cycle of chemotherapy were also found compared with the non-respond group. ROC analyses showed the K value acquired before treatment had the highest diagnostic performance (0.746) in distinguishing responders from non-responders. Furthermore, the high sensitivity (100%) and accuracy (76.3%) from the K value before treatment was found in assessing the response of colorectal liver metastasis to chemotherapy. CONCLUSIONS: The non-mono-exponential diffusion models may be able to predict early response to chemotherapy in patients with colorectal liver metastasis.

Pulmonary nodule segmentation with CT sample synthesis using adversarial networks.

PURPOSE: Segmentation of pulmonary nodules is critical for the analysis of nodules and lung cancer diagnosis. We present a novel framework of segmentation for various types of nodules using convolutional neural networks (CNNs). METHODS: The proposed framework is composed of two major parts. The first part is to increase the variety of samples and build a more balanced dataset. A conditional generative adversarial network (cGAN) is employed to produce synthetic CT images. Semantic labels are generated to impart spatial contextual knowledge to the network. Nine attribute scoring labels are combined as well to preserve nodule features. To refine the realism of synthesized samples, reconstruction error loss is introduced into cGAN. The second part is to train a nodule segmentation network on the extended dataset. We build a three-dimensional (3D) CNN model that exploits heterogeneous maps including edge maps and local binary pattern maps. The incorporation of these maps informs the model of texture patterns and boundary information of nodules, which assists high-level feature learning for segmentation. Residual unit, which learns to reduce residual error, is adopted to accelerate training and improve accuracy. RESULTS: Validation on LIDC-IDRI dataset demonstrates that the generated samples are realistic. The mean squared error and average cosine similarity between real and synthesized samples are 1.55 × 10 - 2 and 0.9534, respectively. The Dice coefficient, positive predicted value, sensitivity, and accuracy are, respectively, 0.8483, 0.8895, 0.8511, and 0.9904 for the segmentation results. CONCLUSIONS: The proposed 3D CNN segmentation framework, based on the use of synthesized samples and multiple maps with residual learning, achieves more accurate nodule segmentation compared to existing state-of-the-art methods. The proposed CT image synthesis method can not only output samples close to real images but also allow for stochastic variation in image diversity.

Evaluation of non-monoexponential diffusion models for hepatocellular carcinoma using b values up to 2000 s/mm2 : A short-term repeatability study.

BACKGROUND: Non-monoexponential diffusion models are being used increasingly for the characterization and curative effect evaluation of hepatocellular carcinoma (HCC). But the fitting quality of the models and the repeatability of their parameters have not been assessed for HCC. PURPOSE: To evaluate kurtosis, stretched exponential, and statistical models for diffusion-weighted imaging (DWI) of HCC, using b-values up to 2000 s/mm2 , in terms of fitting quality and repeatability. STUDY TYPE: Prospective. POPULATION: Eighteen patients with HCC. FIELD STRENGTH/SEQUENCE: Conventional and DW images (b = 0, 200, 500, 1000, 1500, 2000 s/mm2 ) were acquired at 3.0T. ASSESSMENT: The parameters of the kurtosis, stretched exponential, and statistical models were calculated on regions of interest (ROIs) of each lesion. STATISTICAL TESTS: The fitting quality was evaluated through comparing the fitting residuals produced on the average data of ROI between different models using a paired t-test or Wilcoxon rank-sum test. Repeatability of the fitted parameters at the median values on the voxelwise data of ROI was assessed using the within coefficient of variation (WCV), the intraclass correlation coefficient (ICC), and the 95% Bland-Altman limits of agreements (BA-LA). The repeatability was divided into four levels: excellent, good, acceptable, and poor, referring to the values of ICC and WCV. RESULTS: Among three models, the stretched exponential model provided the best fit to HCC (P < 0.05), whereas the statistical model produced the largest fitting residuals (P < 0.05). The repeatability of K from the kurtosis model was excellent (ICC 0.915; WCV 8.79%), while the distributed diffusion coefficient (DDC) from the stretched model was just acceptable (ICC 0.477; WCV 27.83%). The repeatability was good for other diffusion-related parameters. DATA CONCLUSION: Considering the model fit and repeatability, the kurtosis and stretched exponential models are the preferred models for the description of the DW signals of HCC with respect to the statistical model. LEVEL OF EVIDENCE: 2 Technical Efficacy: Stage 2 J. Magn. Reson. Imaging 2019;50:297-304.

Determination of Hepatocellular Carcinoma and Characterization of Hepatic Focal Lesions with Adaptive Multi-Exponential Intravoxel Incoherent Motion Model.

PURPOSE: To distinguish hepatocellular carcinoma (HCC) from other types of hepatic lesions with the adaptive multi-exponential IVIM model. METHODS: 94 hepatic focal lesions, including 38 HCC, 16 metastasis, 12 focal nodular hyperplasia, 13 cholangiocarcinoma, and 15 hemangioma, were examined in this study. Diffusion-weighted images were acquired with 13 b values (b = 0, 3, …, 500 s/mm2) to measure the adaptive multi-exponential IVIM parameters, namely, pure diffusion coefficient (D), diffusion fraction (fd), pseudo-diffusion coefficient (Di*) and perfusion-related diffusion fraction (fi) of the ith perfusion component. Comparison of the parameters of and their diagnostic performance was determined using Mann-Whitney U test, independent-sample t test, one-way analysis of variance, Z test and receiver-operating characteristic analysis. RESULTS: D, D1* and D2* presented significantly difference between HCCs and other hepatic lesions, whereas fd, f1 and f2 did not show statistical differences. In the differential diagnosis of HCCs from other hepatic lesions, D2* (AUC, 0.927) provided best diagnostic performance among all parameters. Additionally, the number of exponential terms in the model was also an important indicator for distinguishing HCCs from other hepatic lesions. In the benign and malignant analysis, D gave the greatest AUC values, 0.895 or 0.853, for differentiation between malignant and benign lesions with three or two exponential terms. Most parameters were not significantly different between hypovascular and hypervascular lesions. For multiple comparisons, significant differences of D, D1* or D2* were found between certain lesion types. CONCLUSION: The adaptive multi-exponential IVIM model was useful and reliable to distinguish HCC from other hepatic lesions.